Rare DNA rearrangements that require no extended sequence similarity (of parental molecules) can cause severe genetic alterations of chromosomal fragments in somatic cells. Thus far about 600 parental sequences for this kind of crossing-over are known. It is expected that their number will increase with the rate 50-100 per year. The purpose of this new project is to store available data on illegitimate recombination in one place. Another goal is to provide simple tools for experimental molecular biologists to decide if newly sequenced DNA fragments contain potential recombination 'hot-spots' or not. The research part of this project concerns analysis of known sequences and possible elucidation of mechanisms for non-homologous recombination. Methods of sequence analysis and molecular modeling have been applied thus far.